I'm just thinking out loud here with hopes of getting opinions on this.
The belief that chess players see moves three or four moves into the future is misleading. In fact chess players do not see the future, but possible good replies. If the road looks like it will result in the best better position with all possible good replies, then take it.
Of course plans are reevaluated after each of the opponents moves.
Instead of seeing the moves, one simply manages the board based on probable moves which if both players know good moves can be fairly predictable at least in the near future.
Originally posted by @eladar I'm just thinking out loud here with hopes of getting opinions on this.
The belief that chess players see moves three or four moves into the future is misleading. In fact chess players do not see the future, but possible good replies. If the road looks like it will result in the best better position with all possible good replies, then take it.
Of cour ...[text shortened]... ves which if both players know good moves can be fairly predictable at least in the near future.
"How many moves ahead can you see?" is a question I get from non-chess players after I tell them that chess is one of my interests.
Human players think heuristically: we're not going to bother calculating lines of play that do not pertain to some specific goal - weakening of the enemy King position/pawn structure, or gaining of space or material, etc.
Even chess engines have had to modify the pure brute-force approach to attain the strength they have. Now, there are algorithms such as "futility-pruning" which tells the machine to stop calculating a line if the evaluation keeps getting worse and worse. Also, material is no longer the only number in the evaluation. Things such as King safety are factored in.
Originally posted by @bigdoggproblem "How many moves ahead can you see?" is a question I get from non-chess players after I tell them that chess is one of my interests.
Human players think heuristically: we're not going to bother calculating lines of play that do not pertain to some specific goal - weakening of the enemy King position/pawn structure, or gaining of space or material, ...[text shortened]... rial is no longer the only number in the evaluation. Things such as King safety are factored in.
When you calculate, how many moves do you try to see? In general of course.
Originally posted by @eladar I'm just thinking out loud here with hopes of getting opinions on this.
The belief that chess players see moves three or four moves into the future is misleading. In fact chess players do not see the future, but possible good replies. If the road looks like it will result in the best better position with all possible good replies, then take it.
Of cour ...[text shortened]... ves which if both players know good moves can be fairly predictable at least in the near future.
Instead of seeing the moves, one simply manages the board based on probable moves which if both players know good moves can be fairly predictable at least in the near future.
That's possible I suppose, but in talking with a number of fairly strong players over the last 30 yrs. (2100 - 2500) "Seeing What Is Generally Good" more often involves studying all levels of tactics, learning how to recognize these tactical patterns, and how to use them to their maximum benefit. This is a heavy lift, and involves consistent study mixed with consistent tournament play. Just one more reason why geeks like me never become really good! 😛
Originally posted by @mchill Instead of seeing the moves, one simply manages the board based on probable moves which if both players know good moves can be fairly predictable at least in the near future.
That's possible I suppose, but in talking with a number of fairly strong players over the last 30 yrs. (2100 - 2500) "Seeing What Is Generally Good" more often involves st ...[text shortened]... consistent tournament play. Just one more reason why geeks like me never become really good! 😛
Is it 300 memorized games required to be a GM? Lots of pattern recognition, which means less actual thinking.
When I first started playing people told me to put my pieces on good squares. I asked which were the good ones, the black or the white.
Originally posted by @eladar Is it 300 memorized games required to be a GM? Lots of pattern recognition, which means less actual thinking.
When I first started playing people told me to put my pieces on good squares. I asked which were the good ones, the black or the white.
I guess you could do that, however I was thinking more in terms of these two: The first, covers all the basics such as the use of pins, forks, deflection, skewers etc. then progresses to intermediate combinations, and how see them more clearly and make them work for you. Averbakh's book is serious workman like tactics study that will challenge even masters. JMHO - These 2 books are the kind of study one needs if they really want to improve.
Originally posted by @eladar When you calculate, how many moves do you try to see? In general of course.
As many as necessary and feasible, which depends on the position. In a forced line, as far as I can manage. When I'm just getting my pieces into position, just the one, only to see if there are any obvious traps I've overlooked. In general? I'm not even sure a general position exists.
Mind you, if you were to ask how many positions I succeed in calculating, the answer would be lower... sometimes zero :/
Originally posted by @eladar Is it 300 memorized games required to be a GM? Lots of pattern recognition, which means less actual thinking.
Required? Yes. Sufficient? Not nearly. More like 30,000, afaict.
Note that "memorised", in this case, does not imply rote learning. It means they analyse and replay them so often they internalise them through familiarity.
Originally posted by @eladar When you calculate, how many moves do you try to see? In general of course.
3-4 in forcing situations. More moves in simple endgames, where the lack of material makes it possible to do so accurately. If there aren't obvious tactical concerns, I focus more on making a plan.
Originally posted by @bigdoggproblem 3-4 in forcing situations. More moves in simple endgames, where the lack of material makes it possible to do so accurately. If there aren't obvious tactical concerns, I focus more on making a plan.
Making a plan. Great topic.
I wish I could make a plan. I feel that I simply hope that my opponent screws up.
Originally posted by @eladar Making a plan. Great topic. ...
Agreed. There is the basic stuff, like developing your pieces, controlling the center, protecting the king. I find myself focusing on stuff I think may lead to an advantage. For example, maybe I see that my pawn structure is inhibiting one of your bishops. I'll try to maximize that effect, often to the detriment of a bunch of other stuff, and I get blindsided.
I guess I'm saying that chess plans have to be malleable.
Originally posted by @apathist Agreed. There is the basic stuff, like developing your pieces, controlling the center, protecting the king. I find myself focusing on stuff I think may lead to an advantage. For example, maybe I see that my pawn structure is inhibiting one of your bishops. I'll try to maximize that effect, often to the detriment of a bunch of other stuff, and I get blindsided.
I guess I'm saying that chess plans have to be malleable.
In other words, your view of what is Generally Good is stagnant while the board is ever changing.
Seeing What is Generally Good
22 Jul '17 17:58
Back to Top